Corrosion Behavior of Passivated Martensitic and Semi-Austenitic Precipitation Hardening Stainless Steel

: This research aimed to conduct a passive layer state study on martensitic and semi-austenitic precipitation hardening stainless steels (PHSS) passivated in citric acid and nitric acid baths at 49 and 70 ◦ C for 50 and 75 min and subsequently exposed in 5 wt.% NaCl and 1 wt.% H 2 SO 4 solutions. Corrosion behavior of the passivated material was observed by using potentiodynamic polarization (PP) according to the ASTM G5-11 standard. The microstructural analysis was performed by optical microscopy and scanning electron microscopy (SEM), while the passivated layer was characterized by X-ray photoelectron spectroscopy (XPS). The results indicated that the semi-austenitic-NA-50 min-70 ◦ C sample showed the best corrosion resistance behavior in both solutions. The XPS characterization conﬁrmed that the martensitic and semi-austenitic surface ﬁlm presented a mixture of chemical compounds, such as Cr 2 O 3 and Fe(OH)O, respectively.


Introduction
Corrosion deterioration of materials in the aircraft industry is a major problem affecting economic, safety, and logistical issues. Corrosion protection methods seek to be friendly to the environment due to the demands in the aeronautical sector. Passivation treatment is a protection method commonly used in stainless steels (SS) to increase their corrosion resistance [1][2][3][4]. Stainless steels are widely used in the industry due to their excellent mechanical resistance, corrosion resistance, and impact resistance.
According to their microstructures, stainless steels are classified into ferritic (α), martensitic (α'), austenitic (γ), duplex (mixture of ferrite and austenite), and precipitation hardening (PH) [1,5,6]. Three main types are typically used in the aeronautical industry: austenitic, martensitic, and PH stainless steel. Good performance in aggressive environments has been observed due to their corrosion, mechanical, and high-temperature properties. They can be used in aircraft components such as actuators, fasteners, and landing gear supports [6][7][8]. Aeronautical materials such as stainless steel are used in In 2004, El-Taib Heakal et al. [33] concluded that the effect of pH on the electrochemical behavior of austenitic stainless steels in naturally aerated and nitrogen-deaerated buffer solutions was studied. Potentiodynamic polarization and electrochemical impedance spectroscopy techniques were used. The results indicate that corrosion current (i corr ) decreases with pH due to an effect of alloying elements such as molybdenum. Ameer et al. [34] reported potentiodynamic polarization results showed that i corr and i c increase with increasing either Cl − or SO 4 2− concentration, indicating the decrease in the passivity of the formed film. The stability of naturally grown passive films on Mo-containing stainless steel was studied by El-Taib Heakal et al. [35] in aerated and deaerated buffer solutions with different pH and in sulfate and chloride solutions. Analysis of the EIS data indicates that the total resistance (RT) of the passive film has higher values in aerated solutions and is generally lower in basic solutions. This indicates that lower solution pH favors the formation of oxide films, offering better corrosion resistance. The higher values of RT in Na 2 SO 4 solutions suggest the formation of more stable passive films in sulfate than in chloride solutions.
This work aimed to study the passive state of martensitic and semi-austenitic PHSS passivated in citric and nitric acid baths at 49 and 70 • C for 50 and 75 min, immersed in 5 wt.% NaCl and 1 wt.% H 2 SO 4 solutions, applying potentiodynamic polarization (PP). The microstructural analysis was performed by scanning electron microscopy (SEM) and optical microscopy (OM). The passivated layer was characterized by X-ray photoelectron spectroscopy (XPS).

Materials
In the current investigation, the martensitic and semi-austenitic PHSS steels were employed and tested in the as-received condition. The nominal chemical composition of these PHSS [47,48] is shown in Table 1. PHSS samples were prepared by metallography (grinding and polishing employed 400, 500, 600, and 800 grade SiC sandpaper) and, subsequently, cleaned for 10 min using ultrasonic in ethanol and deionized water [49].
A design of experiments 5 (DoE) was carried out; due to different combinations of acid solutions, applied with 3-factor; and 2-tier to obtain the optimal concentration, temperature, and passivation time in the baths. Table 2 shows the passivation treatment parameters exposure for each type of PHSS. A design of experiments 5 (DoE) was carried out; due to different combinations of acid solutions, applied with 3-factor; and 2-tier to obtain the optimal concentration, temperature, and passivation time in the baths. Table 2 shows the passivation treatment parameters exposure for each type of PHSS.

Microstructural Characterization
Optical microscopy (OM, Olympus, Hamburg, Germany) was used to determine the microstructure of PHSS; while the micrographs were taken by scanning electron microscopy (SEM, JEOL-JSM-5610LV, Tokyo, Japan) using a secondary electrons (SE) detector at 500×, operating at 20 kV, WD = 14 mm.

Corrosion Test
Corrosion measurements were conducted at room temperature using an electrochemical interface mod. 1287A-Solartron (Bognor Regis, UK) two 5 wt.% NaCl and 1 wt.% H2SO4 solutions [5,6]. The corrosion cell configuration consisted of a working electrode, WE (material to study), a reference electrode of saturated calomel (SCE), and a platinum mesh that served as a counter electrode (CE) for the current measurements [6,51,52]. The potentiodynamic polarization (PP) parameters regarding the potential scan range were applied between -1.0 and 1.0 V from OCP and a sweep rate of 0.06 V/min, according to ASTM G5-11 [50,53]. Tests were realized in triplicate.

Microstructural Characterization
Optical microscopy (OM, Olympus, Hamburg, Germany) was used to determine the microstructure of PHSS; while the micrographs were taken by scanning electron microscopy (SEM, JEOL-JSM-5610LV, Tokyo, Japan) using a secondary electrons (SE) detector at 500×, operating at 20 kV, WD = 14 mm.

Corrosion Test
Corrosion measurements were conducted at room temperature using an electrochemical interface mod. 1287A-Solartron (Bognor Regis, UK) two 5 wt.% NaCl and 1 wt.% H 2 SO 4 solutions [5,6]. The corrosion cell configuration consisted of a working electrode, WE (material to study), a reference electrode of saturated calomel (SCE), and a platinum mesh that served as a counter electrode (CE) for the current measurements [6,51,52]. The potentiodynamic polarization (PP) parameters regarding the potential scan range were applied between −1.0 and 1.0 V from OCP and a sweep rate of 0.06 V/min, according to ASTM G5-11 [50,53]. Tests were realized in triplicate.

XPS Characterization
High-resolution XPS analyses determined the oxide layer's surface chemical composition and valence states in the martensitic and semi-austenitic PHSS. A Thermo Fisher Scientific ESCALAB 250 Xi equipment (Waltham, MA, USA) was operated at a pressure of 10 mBar; the analysis conditions for the high-resolution zones and analysis radius of µm, (eV step energy, 45 • of "take-off angle" with 0.1 eV step), and the excitation of the analyzed photoelectrons were measured with a monochromatic Al KAlpha X-ray source (1486 eV).

OM-SEM Microstructural
OM and SEM techniques were applied to study the microstructures of the samples in initial conditions. Figure 2 shows the martensitic and semi-austenitic PHSS micrographs (a, b) OM and (a', b') SEM-SE. The martensitic PHSS shows a martensitic (α') phase, and semi-austenitic PHSS contains a microstructure of austenite (γ) and delta (δ) ferrite phase, respectively. The SEM analysis corroborates the results. The presence of austenite (γ) in semi-austenitic steels is a thermodynamically stable phase, where an aging treatment can transform this phase. Likewise, alloying elements were found to indicate the presence of the delta ferrite (δ) phase [54][55][56].
High-resolution XPS analyses determined the oxide layer's surface chemical composition and valence states in the martensitic and semi-austenitic PHSS. A Thermo Fisher Scientific ESCALAB 250 Xi equipment (Waltham, MA, USA) was operated at a pressure of 10 mBar; the analysis conditions for the high-resolution zones and analysis radius of µm, (eV step energy, 45° of "take-off angle" with 0.1 eV step), and the excitation of the analyzed photoelectrons were measured with a monochromatic Al KAlpha X-ray source (1486 eV).

OM-SEM Microstructural
OM and SEM techniques were applied to study the microstructures of the samples in initial conditions. Figure 2 shows the martensitic and semi-austenitic PHSS micrographs (a, b) OM and (a', b') SEM-SE. The martensitic PHSS shows a martensitic (α') phase, and semi-austenitic PHSS contains a microstructure of austenite (γ) and delta (δ) ferrite phase, respectively. The SEM analysis corroborates the results. The presence of austenite (γ) in semi-austenitic steels is a thermodynamically stable phase, where an aging treatment can transform this phase. Likewise, alloying elements were found to indicate the presence of the delta ferrite (δ) phase [54][55][56]. The martensitic stainless steels have higher carbon contents than most that are semiaustenitic. This reduces the corrosion resistance but increases mechanical properties such as toughness and increases the susceptibility to chromium carbide precipitation at grain boundaries.
The pitting corrosion is often measured using the pitting resistance equivalent number (PREN). From this theoretical standpoint, and based on their chemical compositions, The martensitic stainless steels have higher carbon contents than most that are semiaustenitic. This reduces the corrosion resistance but increases mechanical properties such as toughness and increases the susceptibility to chromium carbide precipitation at grain boundaries.
The pitting corrosion is often measured using the pitting resistance equivalent number (PREN). From this theoretical standpoint, and based on their chemical compositions, the pitting corrosion resistance of various grades of stainless steels can be compared [57,58]. High PREN values indicate higher corrosion resistance (see Table 3). The PREN is determined by the alloying elements such as chromium, molybdenum, and nitrogen contents, and the most commonly used version of the Equation (1) is: (1)

Corrosion Test Potentiodynamic Polarization
The corrosion behavior was observed by using potentiodynamic polarization. The Tafel extrapolation of PP was employed to determine the corrosion current density, i corr (µA·cm −2 ), potential corrosion, E corr (V), and corrosion rate [5,[59][60][61][62]. The presence of a linear section in the potentiodynamic polarization curve is necessary for using the Tafel extrapolation method. A range of ±300 mV on E corr was determined to be in the linear section of at least one decade of current [63][64][65].   Figure 4 shows the results for PHSS immersed in H 2 SO 4 . The corrosion potentials (E corr ) were found in the range of −0.266 and −0.349 V, respectively. Pitting potential (E pitt ) values were observed from 0.766 up to 0.875 V. All steels under study had a higher and more defined passivation range (0.692 to 1.055 V) than the samples immersed in NaCl. The corrosion rates are higher (×10 −2 mm/yr) except for the semi-austenitic-NA-50 min-70 • C sample that presented a low corrosion rate (3.62 × 10 −4 mm/yr). Tables 4 and 5 show the electrochemical parameters obtained by potentiodynamic polarization (PP) curves. High values of corrosion rates in PHSS immersed in 5 wt.% NaCl solution (within the same order of magnitude) were recorded for nitric acid (×10 −4 mm/yr) and citric acid (×10 −3 mm/yr) passivation. Low values of corrosion rate in PHSS immersed in 1 wt.% H 2 SO 4 solution (in the order of ×10 −2 mm/yr) were recorded, except for the semi-austenitic-NA-50 min-70 • C sample, which presented a low corrosion rate (3.62 × 10 −4 mm/yr). Passivation current and i pass , values are higher for samples exposed to sulfuric acid, ranging from 0.506 up to 9.837 µA·cm −2 .     The current results indicate that all systems showed a mixed activation and passivation followed by a transpassivation or secondary passivation trend.
The transpassive region is above 300 mV vs. ECS for samples in NaCl solution and is above 0.700 V for samples in H 2 SO 4 solution. The passive film formed with nitric and citric acid baths at longer times has higher passivation ranges in H 2 SO 4 than in NaCl solution.

XPS Analysis
XPS measurements determine the valence states, oxide layer composition, and surface chemical in the passivated martensitic and semi-austenitic PHSS, as shown in Figures 5-11.
XPS analysis was conducted with the Avantage software (Waltham, MA, USA). The NIST database (National Institute of Standards and Technology; Gaithersburg, MD, USA) was used to perform the deconvolution calculations of the XPS spectra and assign the chemical compounds through the binding energy peaks.
High-resolution XPS spectra are shown in Figures 5-11, obtained for Cr 2p and O 1s. In XPS spectra for Cr 2p, four chemical compounds were found: Cr 2 O 3 , Cr 7 C 3 , CrO 3 and Cr(C 6 H 6 ) 2, as shown in Figures 5a, 6a, 7a, 8a, 9a, 10a and 11a. According to the literature [88][89][90], the Cr 2 O 3 presence corresponding to the passive layer composition can be found from 576 to 578 eV; the obtained binding energy peaks ranged from 576.69 to 577.07 eV, according to the values reported for the chromium oxide layers. Finally, Figures 5a', 6a', 7a', 8a', 9a', 10a' and 11a' summarize the parameters obtained from full width at half maximum (FWHM) fitting, area ratio, and the peak binding energy, respectively.
In XPS spectra, the contributions of O 1s indicate binding energies for O 1s, from 532.23 to 532.94 eV according to the Gauss-Lorentz peak (see Figures 5b', 6b', 7b', 8b', 9b', 10b' and 11b'). Five chemical species were found Cr 2 O 3 , Fe(OH)O, Fe 2 O 3 , SiO 2 y Cr(OH) 3 . Natajaran et al. [91] indicated that the presence of oxides gives rise to chrome hydroxides and oxide formation on the surface due to OH − . The binding energy peaks obtained for Fe(OH)O and Fe 2 O 3 ranged from 530.26 to 531.08 eV and 528.9 to 531.8 eV, respectively.
The detection of silicon oxide (SiO 2 ) in the high-resolution XPS spectra of the O 1s orbital, whose binding energy is between 532.23 and 532.94 eV, Figures 5b', 7b' and 11b', can be attributed to polishing or impurities in the passivating solutions. In the literature, different species have been found in the oxides generated by the passivation treatment of aged solutions [40,92].
According to the literature, the formation of the chromium oxide layer increases corrosion resistance thanks to the anticorrosive properties provided by chromium oxide.
The analysis of the O 1s contributions in most of the deconvolutions shows different Fe oxides and hydroxides, mainly associated with Fe 3+ . The contributions by Cr 2p show the main peak around 576 eV. As the XPS spectra appear with an inclination at an energy value above 577 eV, the appearance of hydroxide and trioxide species will be more frequent. When analyzing Cr 2p, however, the main peaks were found at an energy level of 576 eV, presumably due to the chromium oxide layer derived from the passivation process.
According to Jung and Mesquita [92,93], there is a relationship between Fe oxides, hydroxides, and Cr oxides. The more Fe is dissolved during the passivation process, the more chromium oxides tend to enrich the passive layer, increasing the corrosion resistance.
Previous studies have revealed that passive films have a bilayer structure [94][95][96]. The compact inner layer is mainly composed of chromium(III) oxide, while the porous outer layer is mainly composed of iron and chromium oxides and hydroxides.

Conclusions
This work studies the passive state of martensitic and semi-austenitic PHSS passivated in citric and nitric acid baths at 49 and 70 °C for 50 and 75 min and immersed in 5 wt.% NaCl and 1 wt.% H2SO4 solutions. From the current experimental results, it can be concluded the following: • OM-SEM characterization indicated that the martensitic PHSS presented a microstructure with a martensitic (α') phase and a semi-austenitic PHSS containing a microstructure of austenite (γ) and delta (δ) ferrite phases, respectively. Based on the values obtained from PREN, the semi-austenitic PHSS (25.37) presented a higher corrosion resistance than the martensitic PHSS (17.26).

•
Potentiodynamic polarization results indicated that martensitic and semi-austenitic steels passivated in nitric acid showed lower corrosion resistance values (in the order of ×10 -4 mm/yr).

•
Nitric acid passivation made the surface susceptible to localized corrosion. The potentiodynamic polarization curves for PHSS immersed in 5 wt.% NaCl solution indicated that the passivation showed a trend since it is not fully defined.

•
Despite having a well-defined passivation layer, passive samples studied in H2SO4 solution presented an increase in corrosion kinetics. • Passivation current, ipass, was found to have higher values for samples exposed to sulfuric acid (from 0.506 up to 9.837 µA·cm −2 ). • XPS analysis determined that the different chemical species on the surface film of martensitic and semi-austenitic PHSS in this work included Cr2O3 and Fe(OH)O. Passive films contained iron and chromium oxide and hydroxide.

•
The samples passivated at 70 °C are the ones that presented the best results independently of the steel, passivating solution, and time.

•
The citric acid passivation process on PHSS could be a green alternative to the currently employed nitric acid passivation process.

Conclusions
This work studies the passive state of martensitic and semi-austenitic PHSS passivated in citric and nitric acid baths at 49 and 70 • C for 50 and 75 min and immersed in 5 wt.% NaCl and 1 wt.% H 2 SO 4 solutions. From the current experimental results, it can be concluded the following: • OM-SEM characterization indicated that the martensitic PHSS presented a microstructure with a martensitic (α') phase and a semi-austenitic PHSS containing a microstructure of austenite (γ) and delta (δ) ferrite phases, respectively. Based on the values obtained from PREN, the semi-austenitic PHSS (25.37) presented a higher corrosion resistance than the martensitic PHSS (17.26).

•
Potentiodynamic polarization results indicated that martensitic and semi-austenitic steels passivated in nitric acid showed lower corrosion resistance values (in the order of ×10 −4 mm/yr). • Nitric acid passivation made the surface susceptible to localized corrosion. The potentiodynamic polarization curves for PHSS immersed in 5 wt.% NaCl solution indicated that the passivation showed a trend since it is not fully defined.

•
Despite having a well-defined passivation layer, passive samples studied in H 2 SO 4 solution presented an increase in corrosion kinetics. • Passivation current, i pass , was found to have higher values for samples exposed to sulfuric acid (from 0.506 up to 9.837 µA·cm −2 ). • XPS analysis determined that the different chemical species on the surface film of martensitic and semi-austenitic PHSS in this work included Cr 2 O 3 and Fe(OH)O. Passive films contained iron and chromium oxide and hydroxide.

•
The samples passivated at 70 • C are the ones that presented the best results independently of the steel, passivating solution, and time.

•
The citric acid passivation process on PHSS could be a green alternative to the currently employed nitric acid passivation process. Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.